checkpoint-resume

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Checkpoint Resume

检查点恢复

Rate-limit-resilient pipeline orchestrator. Saves progress to

.claude/pipeline-state.json

after every phase so long sessions survive interruptions.

具备抗速率限制能力的管道编排器。每个阶段结束后会将进度保存到

.claude/pipeline-state.json

文件中，确保长时会话在中断后可以恢复。

Quick Reference

快速参考

Category	Rule	Impact	Key Pattern
Phase Ordering	`${CLAUDE_SKILL_DIR}/rules/ordering-priority.md`	CRITICAL	GitHub issues/commits first, file-heavy phases last
State Writes	`${CLAUDE_SKILL_DIR}/rules/state-write-timing.md`	CRITICAL	Write after every phase, never batch
Mini-Commits	`${CLAUDE_SKILL_DIR}/rules/checkpoint-mini-commit.md`	HIGH	Every 3 phases, checkpoint commit format

Total: 3 rules across 3 categories

类别	规则	影响级别	核心模式
阶段排序	`${CLAUDE_SKILL_DIR}/rules/ordering-priority.md`	关键	优先处理GitHub Issues/提交，最后处理文件密集型阶段
状态写入	`${CLAUDE_SKILL_DIR}/rules/state-write-timing.md`	关键	每个阶段结束后立即写入，绝不批量处理
小型提交	`${CLAUDE_SKILL_DIR}/rules/checkpoint-mini-commit.md`	高	每3个阶段执行一次检查点格式的小型提交

总计：3个类别下的3条规则

On Invocation

调用时处理

If
.claude/pipeline-state.json
exists: run

scripts/show-status.sh

to display progress, then ask to resume, pick a different phase, or restart. Load

Read("${CLAUDE_SKILL_DIR}/references/resume-decision-tree.md")

for the full decision tree.

If no state file exists: ask the user to describe the task, build an execution plan, write initial state via

scripts/init-pipeline.sh <branch>

, begin Phase 1.

若
.claude/pipeline-state.json
文件存在：运行

scripts/show-status.sh

以展示进度，随后询问用户是恢复任务、选择其他阶段还是重新开始。加载

Read("${CLAUDE_SKILL_DIR}/references/resume-decision-tree.md")

查看完整决策树。

若状态文件不存在： 请用户描述任务，构建执行计划，通过

scripts/init-pipeline.sh <branch>

写入初始状态，然后开始第一阶段。

Execution Plan Structure

执行计划结构

json

{
  "phases": [
    { "id": "create-issues", "name": "Create GitHub Issues", "dependencies": [], "status": "pending" },
    { "id": "commit-scaffold", "name": "Commit Scaffold", "dependencies": [], "status": "pending" },
    { "id": "write-source", "name": "Write Source Files", "dependencies": ["commit-scaffold"], "status": "pending" }
  ]
}

Phases with empty

dependencies

may run in parallel via Task sub-agents (when they don't share file writes).

json

{
  "phases": [
    { "id": "create-issues", "name": "Create GitHub Issues", "dependencies": [], "status": "pending" },
    { "id": "commit-scaffold", "name": "Commit Scaffold", "dependencies": [], "status": "pending" },
    { "id": "write-source", "name": "Write Source Files", "dependencies": ["commit-scaffold"], "status": "pending" }
  ]
}

依赖项为空的阶段可通过Task子代理并行执行（当它们不共享文件写入操作时）。

After Each Phase

每个阶段结束后处理

Update

.claude/pipeline-state.json

— see

Read("${CLAUDE_SKILL_DIR}/rules/state-write-timing.md")

Every 3 phases: create a mini-commit — see

Read("${CLAUDE_SKILL_DIR}/rules/checkpoint-mini-commit.md")

更新

.claude/pipeline-state.json

文件 — 详见

Read("${CLAUDE_SKILL_DIR}/rules/state-write-timing.md")

每3个阶段：创建一个小型提交 — 详见

Read("${CLAUDE_SKILL_DIR}/rules/checkpoint-mini-commit.md")

References

参考资料

Load on demand with

Read("${CLAUDE_SKILL_DIR}/references/<file>")

File	Content
`references/pipeline-state-schema.md`	Full field-by-field schema with examples
`references/pipeline-state.schema.json`	Machine-readable JSON Schema for validation
`references/resume-decision-tree.md`	Logic for resuming, picking phases, or restarting

按需通过

Read("${CLAUDE_SKILL_DIR}/references/<file>")

加载：

文件	内容
`references/pipeline-state-schema.md`	带示例的完整逐字段 schema 说明
`references/pipeline-state.schema.json`	用于验证的机器可读JSON Schema
`references/resume-decision-tree.md`	恢复、阶段选择或重新开始的逻辑说明

Scripts

脚本

```
scripts/init-pipeline.sh <branch>
```
— print skeleton state JSON to stdout
```
scripts/show-status.sh [path]
```
— print human-readable pipeline status (requires
```
jq
```
)

```
scripts/init-pipeline.sh <branch>
```
— 将骨架状态JSON输出到标准输出
```
scripts/show-status.sh [path]
```
— 输出人类可读的管道状态（需要
```
jq
```
依赖）

Key Decisions

核心决策建议

Decision	Recommendation
Phase granularity	One meaningful deliverable per phase (a commit, a set of issues, a feature)
Parallelism	Task sub-agents only for phases with empty `dependencies` that don't share file writes
Rate limit recovery	State is already saved — re-invoke `/checkpoint-resume` to continue

决策项	建议
阶段粒度	每个阶段对应一个有意义的交付成果（一次提交、一组Issues、一个功能）
并行执行	仅当依赖项为空且不共享文件写入操作的阶段，才使用Task子代理并行执行
速率限制恢复	状态已保存 — 重新调用 `/checkpoint-resume` 即可继续任务